Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1652592
| From | Peter Zijlstra <peterz@infradead.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi |
| Date | 2017-05-29 17:30 +0200 |
| Message-ID | <tMuMi-1KX-35@gated-at.bofh.it> (permalink) |
| References | <tLyRY-4YN-13@gated-at.bofh.it> <tMrEK-865-25@gated-at.bofh.it> <tMtdv-A4-13@gated-at.bofh.it> <tMuMi-1KX-31@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
On Mon, May 29, 2017 at 05:22:54PM +0200, Peter Zijlstra wrote:
> On Mon, May 29, 2017 at 04:43:09PM +0300, Alexey Budankov wrote:
> > On 29.05.2017 15:03, Alexander Shishkin wrote:
> > > Alexey Budankov <alexey.budankov@linux.intel.com> writes:
>
> > > > + } else if (event->cpu > node_event->cpu) {
> > > > + node = &((*node)->rb_right);
> > > > + } else {
> > > > + list_add_tail(&event->group_list_entry,
> > > > + &node_event->group_list);
> > >
> > > So why is this better than simply having per-cpu event lists plus one
> > > for per-thread events?
> >
> > Good question. Choice of data structure and layout depends on the operations
> > applied to the data so keeping groups as a tree simplifies and improves the
> > implementation in terms of scalability and performance. Please ask more if
> > any.
>
> Since these lists are per context, and each task can have a context,
> you'd end up with per-task-per-cpu memory, which is something we'd like
> to avoid (some archs have very limited per-cpu memory space etc..).
>
> Also, we'd like to have that tree for other reasons, like for instance
> that heterogeneous PMU crud ARM has. Also, with a tree we can easier do
> time based round-robin scheduling,
>
Oh and in general multi-PMU stuff, aside from hetero PMU becomes much
easier.
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-27 03:40 +0200
[PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-27 13:30 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 09:50 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 11:20 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 12:50 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 13:00 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 13:30 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 13:50 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 09:50 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 11:30 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 12:40 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 12:50 +0200
Re: [PATCH v2]: perf/core: addressing 4x slowdown during per-process, profiling of STREAM benchmark on Intel Xeon Phi David Carrillo-Cisneros <davidcc@google.com> - 2017-05-31 23:40 +0200
Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Alexander Shishkin <alexander.shishkin@linux.intel.com> - 2017-05-29 14:10 +0200
Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 15:50 +0200
Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 17:30 +0200
Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Peter Zijlstra <peterz@infradead.org> - 2017-05-29 17:30 +0200
Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Alexey Budankov <alexey.budankov@linux.intel.com> - 2017-05-29 18:50 +0200
Re: [PATCH]: perf/core: addressing 4x slowdown during per-process profiling of STREAM benchmark on Intel Xeon Phi Alexander Shishkin <alexander.shishkin@linux.intel.com> - 2017-05-30 10:30 +0200
csiph-web